Enhanced Tree Clustering with Single Pronunciation Dictionary for Conversational Speech Rec
- 格式:pdf
- 大小:93.76 KB
- 文档页数:4
新教材2019版人教版新课标高中英语选择性必修2Unit 1cholera[ˈkɒlərə]n霍乱severe[sɪˈvɪə]adj极为恶劣的;十分严重的;严厉的diarrhoea[ˌdaɪəˈrɪə]n腹泻dehydration[ˌdiːhaɪˈdreɪʃən]n脱水frustrated[frʌsˈtreɪtɪd]adj懊恼的;沮丧的;失意的once and for all最终地;彻底地contradictory[ˌkɒntrəˈdɪktəri]adj相互矛盾的;对立的;不一致的infection[ɪnˈfɛkʃən]n感染;传染infect[ɪnˈfɛkt]vt使感染;传染germ[ʤɜːm]n微生物;细菌;病菌subscribe[səbˈskraɪb]vi认购(股份);定期订购;定期交纳(会费)subscribe to同意;赞同proof[pruːf]n证据;证明;检验multiple[ˈmʌltɪpl]adj数量多的;多种多样的pump[pʌmp]n泵;抽水机;打气筒water pump[ˈwɔːtəpʌmp]水泵household[ˈhaʊshəʊld]n一家人;家庭;同住一所(套)房子的人suspect[ˈsʌspɛkt]v t&vi. 怀疑;疑有;不信任。
n犯罪嫌疑人;可疑对象blame[bleɪm]vt把……归咎于;责怪;指责。
n. 责备;指责handle[ˈhændl]n把手;拉手;柄。
vt. 处理;搬动;操纵(车辆、动物、工具等)intervention[ˌɪntə(ː)ˈvɛnʃən]n介入;出面;干涉link[lɪŋk]n联系;纽带。
vt. 把……连接起来;相关联raw[rɔː]adj未煮的;生的;未经处理的;原始的pure[pjʊə]adj干净的;纯的;纯粹的substantial[səbˈstænʃəl]adj大量的;价值巨大的;重大的decrease[ˈdiːkriːs]n减少;降低;减少量。
机器学习专业词汇中英⽂对照activation 激活值activation function 激活函数additive noise 加性噪声autoencoder ⾃编码器Autoencoders ⾃编码算法average firing rate 平均激活率average sum-of-squares error 均⽅差backpropagation 后向传播basis 基basis feature vectors 特征基向量batch gradient ascent 批量梯度上升法Bayesian regularization method 贝叶斯规则化⽅法Bernoulli random variable 伯努利随机变量bias term 偏置项binary classfication ⼆元分类class labels 类型标记concatenation 级联conjugate gradient 共轭梯度contiguous groups 联通区域convex optimization software 凸优化软件convolution 卷积cost function 代价函数covariance matrix 协⽅差矩阵DC component 直流分量decorrelation 去相关degeneracy 退化demensionality reduction 降维derivative 导函数diagonal 对⾓线diffusion of gradients 梯度的弥散eigenvalue 特征值eigenvector 特征向量error term 残差feature matrix 特征矩阵feature standardization 特征标准化feedforward architectures 前馈结构算法feedforward neural network 前馈神经⽹络feedforward pass 前馈传导fine-tuned 微调first-order feature ⼀阶特征forward pass 前向传导forward propagation 前向传播Gaussian prior ⾼斯先验概率generative model ⽣成模型gradient descent 梯度下降Greedy layer-wise training 逐层贪婪训练⽅法grouping matrix 分组矩阵Hadamard product 阿达马乘积Hessian matrix Hessian 矩阵hidden layer 隐含层hidden units 隐藏神经元Hierarchical grouping 层次型分组higher-order features 更⾼阶特征highly non-convex optimization problem ⾼度⾮凸的优化问题histogram 直⽅图hyperbolic tangent 双曲正切函数hypothesis 估值,假设identity activation function 恒等激励函数IID 独⽴同分布illumination 照明inactive 抑制independent component analysis 独⽴成份分析input domains 输⼊域input layer 输⼊层intensity 亮度/灰度intercept term 截距KL divergence 相对熵KL divergence KL分散度k-Means K-均值learning rate 学习速率least squares 最⼩⼆乘法linear correspondence 线性响应linear superposition 线性叠加line-search algorithm 线搜索算法local mean subtraction 局部均值消减local optima 局部最优解logistic regression 逻辑回归loss function 损失函数low-pass filtering 低通滤波magnitude 幅值MAP 极⼤后验估计maximum likelihood estimation 极⼤似然估计mean 平均值MFCC Mel 倒频系数multi-class classification 多元分类neural networks 神经⽹络neuron 神经元Newton’s method ⽜顿法non-convex function ⾮凸函数non-linear feature ⾮线性特征norm 范式norm bounded 有界范数norm constrained 范数约束normalization 归⼀化numerical roundoff errors 数值舍⼊误差numerically checking 数值检验numerically reliable 数值计算上稳定object detection 物体检测objective function ⽬标函数off-by-one error 缺位错误orthogonalization 正交化output layer 输出层overall cost function 总体代价函数over-complete basis 超完备基over-fitting 过拟合parts of objects ⽬标的部件part-whole decompostion 部分-整体分解PCA 主元分析penalty term 惩罚因⼦per-example mean subtraction 逐样本均值消减pooling 池化pretrain 预训练principal components analysis 主成份分析quadratic constraints ⼆次约束RBMs 受限Boltzman机reconstruction based models 基于重构的模型reconstruction cost 重建代价reconstruction term 重构项redundant 冗余reflection matrix 反射矩阵regularization 正则化regularization term 正则化项rescaling 缩放robust 鲁棒性run ⾏程second-order feature ⼆阶特征sigmoid activation function S型激励函数significant digits 有效数字singular value 奇异值singular vector 奇异向量smoothed L1 penalty 平滑的L1范数惩罚Smoothed topographic L1 sparsity penalty 平滑地形L1稀疏惩罚函数smoothing 平滑Softmax Regresson Softmax回归sorted in decreasing order 降序排列source features 源特征sparse autoencoder 消减归⼀化Sparsity 稀疏性sparsity parameter 稀疏性参数sparsity penalty 稀疏惩罚square function 平⽅函数squared-error ⽅差stationary 平稳性(不变性)stationary stochastic process 平稳随机过程step-size 步长值supervised learning 监督学习symmetric positive semi-definite matrix 对称半正定矩阵symmetry breaking 对称失效tanh function 双曲正切函数the average activation 平均活跃度the derivative checking method 梯度验证⽅法the empirical distribution 经验分布函数the energy function 能量函数the Lagrange dual 拉格朗⽇对偶函数the log likelihood 对数似然函数the pixel intensity value 像素灰度值the rate of convergence 收敛速度topographic cost term 拓扑代价项topographic ordered 拓扑秩序transformation 变换translation invariant 平移不变性trivial answer 平凡解under-complete basis 不完备基unrolling 组合扩展unsupervised learning ⽆监督学习variance ⽅差vecotrized implementation 向量化实现vectorization ⽮量化visual cortex 视觉⽪层weight decay 权重衰减weighted average 加权平均值whitening ⽩化zero-mean 均值为零Letter AAccumulated error backpropagation 累积误差逆传播Activation Function 激活函数Adaptive Resonance Theory/ART ⾃适应谐振理论Addictive model 加性学习Adversarial Networks 对抗⽹络Affine Layer 仿射层Affinity matrix 亲和矩阵Agent 代理 / 智能体Algorithm 算法Alpha-beta pruning α-β剪枝Anomaly detection 异常检测Approximation 近似Area Under ROC Curve/AUC Roc 曲线下⾯积Artificial General Intelligence/AGI 通⽤⼈⼯智能Artificial Intelligence/AI ⼈⼯智能Association analysis 关联分析Attention mechanism 注意⼒机制Attribute conditional independence assumption 属性条件独⽴性假设Attribute space 属性空间Attribute value 属性值Autoencoder ⾃编码器Automatic speech recognition ⾃动语⾳识别Automatic summarization ⾃动摘要Average gradient 平均梯度Average-Pooling 平均池化Letter BBackpropagation Through Time 通过时间的反向传播Backpropagation/BP 反向传播Base learner 基学习器Base learning algorithm 基学习算法Batch Normalization/BN 批量归⼀化Bayes decision rule 贝叶斯判定准则Bayes Model Averaging/BMA 贝叶斯模型平均Bayes optimal classifier 贝叶斯最优分类器Bayesian decision theory 贝叶斯决策论Bayesian network 贝叶斯⽹络Between-class scatter matrix 类间散度矩阵Bias 偏置 / 偏差Bias-variance decomposition 偏差-⽅差分解Bias-Variance Dilemma 偏差 – ⽅差困境Bi-directional Long-Short Term Memory/Bi-LSTM 双向长短期记忆Binary classification ⼆分类Binomial test ⼆项检验Bi-partition ⼆分法Boltzmann machine 玻尔兹曼机Bootstrap sampling ⾃助采样法/可重复采样/有放回采样Bootstrapping ⾃助法Break-Event Point/BEP 平衡点Letter CCalibration 校准Cascade-Correlation 级联相关Categorical attribute 离散属性Class-conditional probability 类条件概率Classification and regression tree/CART 分类与回归树Classifier 分类器Class-imbalance 类别不平衡Closed -form 闭式Cluster 簇/类/集群Cluster analysis 聚类分析Clustering 聚类Clustering ensemble 聚类集成Co-adapting 共适应Coding matrix 编码矩阵COLT 国际学习理论会议Committee-based learning 基于委员会的学习Competitive learning 竞争型学习Component learner 组件学习器Comprehensibility 可解释性Computation Cost 计算成本Computational Linguistics 计算语⾔学Computer vision 计算机视觉Concept drift 概念漂移Concept Learning System /CLS 概念学习系统Conditional entropy 条件熵Conditional mutual information 条件互信息Conditional Probability Table/CPT 条件概率表Conditional random field/CRF 条件随机场Conditional risk 条件风险Confidence 置信度Confusion matrix 混淆矩阵Connection weight 连接权Connectionism 连结主义Consistency ⼀致性/相合性Contingency table 列联表Continuous attribute 连续属性Convergence 收敛Conversational agent 会话智能体Convex quadratic programming 凸⼆次规划Convexity 凸性Convolutional neural network/CNN 卷积神经⽹络Co-occurrence 同现Correlation coefficient 相关系数Cosine similarity 余弦相似度Cost curve 成本曲线Cost Function 成本函数Cost matrix 成本矩阵Cost-sensitive 成本敏感Cross entropy 交叉熵Cross validation 交叉验证Crowdsourcing 众包Curse of dimensionality 维数灾难Cut point 截断点Cutting plane algorithm 割平⾯法Letter DData mining 数据挖掘Data set 数据集Decision Boundary 决策边界Decision stump 决策树桩Decision tree 决策树/判定树Deduction 演绎Deep Belief Network 深度信念⽹络Deep Convolutional Generative Adversarial Network/DCGAN 深度卷积⽣成对抗⽹络Deep learning 深度学习Deep neural network/DNN 深度神经⽹络Deep Q-Learning 深度 Q 学习Deep Q-Network 深度 Q ⽹络Density estimation 密度估计Density-based clustering 密度聚类Differentiable neural computer 可微分神经计算机Dimensionality reduction algorithm 降维算法Directed edge 有向边Disagreement measure 不合度量Discriminative model 判别模型Discriminator 判别器Distance measure 距离度量Distance metric learning 距离度量学习Distribution 分布Divergence 散度Diversity measure 多样性度量/差异性度量Domain adaption 领域⾃适应Downsampling 下采样D-separation (Directed separation)有向分离Dual problem 对偶问题Dummy node 哑结点Dynamic Fusion 动态融合Dynamic programming 动态规划Letter EEigenvalue decomposition 特征值分解Embedding 嵌⼊Emotional analysis 情绪分析Empirical conditional entropy 经验条件熵Empirical entropy 经验熵Empirical error 经验误差Empirical risk 经验风险End-to-End 端到端Energy-based model 基于能量的模型Ensemble learning 集成学习Ensemble pruning 集成修剪Error Correcting Output Codes/ECOC 纠错输出码Error rate 错误率Error-ambiguity decomposition 误差-分歧分解Euclidean distance 欧⽒距离Evolutionary computation 演化计算Expectation-Maximization 期望最⼤化Expected loss 期望损失Exploding Gradient Problem 梯度爆炸问题Exponential loss function 指数损失函数Extreme Learning Machine/ELM 超限学习机Letter FFactorization 因⼦分解False negative 假负类False positive 假正类False Positive Rate/FPR 假正例率Feature engineering 特征⼯程Feature selection 特征选择Feature vector 特征向量Featured Learning 特征学习Feedforward Neural Networks/FNN 前馈神经⽹络Fine-tuning 微调Flipping output 翻转法Fluctuation 震荡Forward stagewise algorithm 前向分步算法Frequentist 频率主义学派Full-rank matrix 满秩矩阵Functional neuron 功能神经元Letter GGain ratio 增益率Game theory 博弈论Gaussian kernel function ⾼斯核函数Gaussian Mixture Model ⾼斯混合模型General Problem Solving 通⽤问题求解Generalization 泛化Generalization error 泛化误差Generalization error bound 泛化误差上界Generalized Lagrange function ⼴义拉格朗⽇函数Generalized linear model ⼴义线性模型Generalized Rayleigh quotient ⼴义瑞利商Generative Adversarial Networks/GAN ⽣成对抗⽹络Generative Model ⽣成模型Generator ⽣成器Genetic Algorithm/GA 遗传算法Gibbs sampling 吉布斯采样Gini index 基尼指数Global minimum 全局最⼩Global Optimization 全局优化Gradient boosting 梯度提升Gradient Descent 梯度下降Graph theory 图论Ground-truth 真相/真实Letter HHard margin 硬间隔Hard voting 硬投票Harmonic mean 调和平均Hesse matrix 海塞矩阵Hidden dynamic model 隐动态模型Hidden layer 隐藏层Hidden Markov Model/HMM 隐马尔可夫模型Hierarchical clustering 层次聚类Hilbert space 希尔伯特空间Hinge loss function 合页损失函数Hold-out 留出法Homogeneous 同质Hybrid computing 混合计算Hyperparameter 超参数Hypothesis 假设Hypothesis test 假设验证Letter IICML 国际机器学习会议Improved iterative scaling/IIS 改进的迭代尺度法Incremental learning 增量学习Independent and identically distributed/i.i.d. 独⽴同分布Independent Component Analysis/ICA 独⽴成分分析Indicator function 指⽰函数Individual learner 个体学习器Induction 归纳Inductive bias 归纳偏好Inductive learning 归纳学习Inductive Logic Programming/ILP 归纳逻辑程序设计Information entropy 信息熵Information gain 信息增益Input layer 输⼊层Insensitive loss 不敏感损失Inter-cluster similarity 簇间相似度International Conference for Machine Learning/ICML 国际机器学习⼤会Intra-cluster similarity 簇内相似度Intrinsic value 固有值Isometric Mapping/Isomap 等度量映射Isotonic regression 等分回归Iterative Dichotomiser 迭代⼆分器Letter KKernel method 核⽅法Kernel trick 核技巧Kernelized Linear Discriminant Analysis/KLDA 核线性判别分析K-fold cross validation k 折交叉验证/k 倍交叉验证K-Means Clustering K – 均值聚类K-Nearest Neighbours Algorithm/KNN K近邻算法Knowledge base 知识库Knowledge Representation 知识表征Letter LLabel space 标记空间Lagrange duality 拉格朗⽇对偶性Lagrange multiplier 拉格朗⽇乘⼦Laplace smoothing 拉普拉斯平滑Laplacian correction 拉普拉斯修正Latent Dirichlet Allocation 隐狄利克雷分布Latent semantic analysis 潜在语义分析Latent variable 隐变量Lazy learning 懒惰学习Learner 学习器Learning by analogy 类⽐学习Learning rate 学习率Learning Vector Quantization/LVQ 学习向量量化Least squares regression tree 最⼩⼆乘回归树Leave-One-Out/LOO 留⼀法linear chain conditional random field 线性链条件随机场Linear Discriminant Analysis/LDA 线性判别分析Linear model 线性模型Linear Regression 线性回归Link function 联系函数Local Markov property 局部马尔可夫性Local minimum 局部最⼩Log likelihood 对数似然Log odds/logit 对数⼏率Logistic Regression Logistic 回归Log-likelihood 对数似然Log-linear regression 对数线性回归Long-Short Term Memory/LSTM 长短期记忆Loss function 损失函数Letter MMachine translation/MT 机器翻译Macron-P 宏查准率Macron-R 宏查全率Majority voting 绝对多数投票法Manifold assumption 流形假设Manifold learning 流形学习Margin theory 间隔理论Marginal distribution 边际分布Marginal independence 边际独⽴性Marginalization 边际化Markov Chain Monte Carlo/MCMC 马尔可夫链蒙特卡罗⽅法Markov Random Field 马尔可夫随机场Maximal clique 最⼤团Maximum Likelihood Estimation/MLE 极⼤似然估计/极⼤似然法Maximum margin 最⼤间隔Maximum weighted spanning tree 最⼤带权⽣成树Max-Pooling 最⼤池化Mean squared error 均⽅误差Meta-learner 元学习器Metric learning 度量学习Micro-P 微查准率Micro-R 微查全率Minimal Description Length/MDL 最⼩描述长度Minimax game 极⼩极⼤博弈Misclassification cost 误分类成本Mixture of experts 混合专家Momentum 动量Moral graph 道德图/端正图Multi-class classification 多分类Multi-document summarization 多⽂档摘要Multi-layer feedforward neural networks 多层前馈神经⽹络Multilayer Perceptron/MLP 多层感知器Multimodal learning 多模态学习Multiple Dimensional Scaling 多维缩放Multiple linear regression 多元线性回归Multi-response Linear Regression /MLR 多响应线性回归Mutual information 互信息Letter NNaive bayes 朴素贝叶斯Naive Bayes Classifier 朴素贝叶斯分类器Named entity recognition 命名实体识别Nash equilibrium 纳什均衡Natural language generation/NLG ⾃然语⾔⽣成Natural language processing ⾃然语⾔处理Negative class 负类Negative correlation 负相关法Negative Log Likelihood 负对数似然Neighbourhood Component Analysis/NCA 近邻成分分析Neural Machine Translation 神经机器翻译Neural Turing Machine 神经图灵机Newton method ⽜顿法NIPS 国际神经信息处理系统会议No Free Lunch Theorem/NFL 没有免费的午餐定理Noise-contrastive estimation 噪⾳对⽐估计Nominal attribute 列名属性Non-convex optimization ⾮凸优化Nonlinear model ⾮线性模型Non-metric distance ⾮度量距离Non-negative matrix factorization ⾮负矩阵分解Non-ordinal attribute ⽆序属性Non-Saturating Game ⾮饱和博弈Norm 范数Normalization 归⼀化Nuclear norm 核范数Numerical attribute 数值属性Letter OObjective function ⽬标函数Oblique decision tree 斜决策树Occam’s razor 奥卡姆剃⼑Odds ⼏率Off-Policy 离策略One shot learning ⼀次性学习One-Dependent Estimator/ODE 独依赖估计On-Policy 在策略Ordinal attribute 有序属性Out-of-bag estimate 包外估计Output layer 输出层Output smearing 输出调制法Overfitting 过拟合/过配Oversampling 过采样Letter PPaired t-test 成对 t 检验Pairwise 成对型Pairwise Markov property 成对马尔可夫性Parameter 参数Parameter estimation 参数估计Parameter tuning 调参Parse tree 解析树Particle Swarm Optimization/PSO 粒⼦群优化算法Part-of-speech tagging 词性标注Perceptron 感知机Performance measure 性能度量Plug and Play Generative Network 即插即⽤⽣成⽹络Plurality voting 相对多数投票法Polarity detection 极性检测Polynomial kernel function 多项式核函数Pooling 池化Positive class 正类Positive definite matrix 正定矩阵Post-hoc test 后续检验Post-pruning 后剪枝potential function 势函数Precision 查准率/准确率Prepruning 预剪枝Principal component analysis/PCA 主成分分析Principle of multiple explanations 多释原则Prior 先验Probability Graphical Model 概率图模型Proximal Gradient Descent/PGD 近端梯度下降Pruning 剪枝Pseudo-label 伪标记Letter QQuantized Neural Network 量⼦化神经⽹络Quantum computer 量⼦计算机Quantum Computing 量⼦计算Quasi Newton method 拟⽜顿法Letter RRadial Basis Function/RBF 径向基函数Random Forest Algorithm 随机森林算法Random walk 随机漫步Recall 查全率/召回率Receiver Operating Characteristic/ROC 受试者⼯作特征Rectified Linear Unit/ReLU 线性修正单元Recurrent Neural Network 循环神经⽹络Recursive neural network 递归神经⽹络Reference model 参考模型Regression 回归Regularization 正则化Reinforcement learning/RL 强化学习Representation learning 表征学习Representer theorem 表⽰定理reproducing kernel Hilbert space/RKHS 再⽣核希尔伯特空间Re-sampling 重采样法Rescaling 再缩放Residual Mapping 残差映射Residual Network 残差⽹络Restricted Boltzmann Machine/RBM 受限玻尔兹曼机Restricted Isometry Property/RIP 限定等距性Re-weighting 重赋权法Robustness 稳健性/鲁棒性Root node 根结点Rule Engine 规则引擎Rule learning 规则学习Letter SSaddle point 鞍点Sample space 样本空间Sampling 采样Score function 评分函数Self-Driving ⾃动驾驶Self-Organizing Map/SOM ⾃组织映射Semi-naive Bayes classifiers 半朴素贝叶斯分类器Semi-Supervised Learning 半监督学习semi-Supervised Support Vector Machine 半监督⽀持向量机Sentiment analysis 情感分析Separating hyperplane 分离超平⾯Sigmoid function Sigmoid 函数Similarity measure 相似度度量Simulated annealing 模拟退⽕Simultaneous localization and mapping 同步定位与地图构建Singular Value Decomposition 奇异值分解Slack variables 松弛变量Smoothing 平滑Soft margin 软间隔Soft margin maximization 软间隔最⼤化Soft voting 软投票Sparse representation 稀疏表征Sparsity 稀疏性Specialization 特化Spectral Clustering 谱聚类Speech Recognition 语⾳识别Splitting variable 切分变量Squashing function 挤压函数Stability-plasticity dilemma 可塑性-稳定性困境Statistical learning 统计学习Status feature function 状态特征函Stochastic gradient descent 随机梯度下降Stratified sampling 分层采样Structural risk 结构风险Structural risk minimization/SRM 结构风险最⼩化Subspace ⼦空间Supervised learning 监督学习/有导师学习support vector expansion ⽀持向量展式Support Vector Machine/SVM ⽀持向量机Surrogat loss 替代损失Surrogate function 替代函数Symbolic learning 符号学习Symbolism 符号主义Synset 同义词集Letter TT-Distribution Stochastic Neighbour Embedding/t-SNE T – 分布随机近邻嵌⼊Tensor 张量Tensor Processing Units/TPU 张量处理单元The least square method 最⼩⼆乘法Threshold 阈值Threshold logic unit 阈值逻辑单元Threshold-moving 阈值移动Time Step 时间步骤Tokenization 标记化Training error 训练误差Training instance 训练⽰例/训练例Transductive learning 直推学习Transfer learning 迁移学习Treebank 树库Tria-by-error 试错法True negative 真负类True positive 真正类True Positive Rate/TPR 真正例率Turing Machine 图灵机Twice-learning ⼆次学习Letter UUnderfitting ⽋拟合/⽋配Undersampling ⽋采样Understandability 可理解性Unequal cost ⾮均等代价Unit-step function 单位阶跃函数Univariate decision tree 单变量决策树Unsupervised learning ⽆监督学习/⽆导师学习Unsupervised layer-wise training ⽆监督逐层训练Upsampling 上采样Letter VVanishing Gradient Problem 梯度消失问题Variational inference 变分推断VC Theory VC维理论Version space 版本空间Viterbi algorithm 维特⽐算法Von Neumann architecture 冯 · 诺伊曼架构Letter WWasserstein GAN/WGAN Wasserstein⽣成对抗⽹络Weak learner 弱学习器Weight 权重Weight sharing 权共享Weighted voting 加权投票法Within-class scatter matrix 类内散度矩阵Word embedding 词嵌⼊Word sense disambiguation 词义消歧Letter ZZero-data learning 零数据学习Zero-shot learning 零次学习Aapproximations近似值arbitrary随意的affine仿射的arbitrary任意的amino acid氨基酸amenable经得起检验的axiom公理,原则abstract提取architecture架构,体系结构;建造业absolute绝对的arsenal军⽕库assignment分配algebra线性代数asymptotically⽆症状的appropriate恰当的Bbias偏差brevity简短,简洁;短暂broader⼴泛briefly简短的batch批量Cconvergence 收敛,集中到⼀点convex凸的contours轮廓constraint约束constant常理commercial商务的complementarity补充coordinate ascent同等级上升clipping剪下物;剪报;修剪component分量;部件continuous连续的covariance协⽅差canonical正规的,正则的concave⾮凸的corresponds相符合;相当;通信corollary推论concrete具体的事物,实在的东西cross validation交叉验证correlation相互关系convention约定cluster⼀簇centroids 质⼼,形⼼converge收敛computationally计算(机)的calculus计算Dderive获得,取得dual⼆元的duality⼆元性;⼆象性;对偶性derivation求导;得到;起源denote预⽰,表⽰,是…的标志;意味着,[逻]指称divergence 散度;发散性dimension尺度,规格;维数dot⼩圆点distortion变形density概率密度函数discrete离散的discriminative有识别能⼒的diagonal对⾓dispersion分散,散开determinant决定因素disjoint不相交的Eencounter遇到ellipses椭圆equality等式extra额外的empirical经验;观察ennmerate例举,计数exceed超过,越出expectation期望efficient⽣效的endow赋予explicitly清楚的exponential family指数家族equivalently等价的Ffeasible可⾏的forary初次尝试finite有限的,限定的forgo摒弃,放弃fliter过滤frequentist最常发⽣的forward search前向式搜索formalize使定形Ggeneralized归纳的generalization概括,归纳;普遍化;判断(根据不⾜)guarantee保证;抵押品generate形成,产⽣geometric margins⼏何边界gap裂⼝generative⽣产的;有⽣产⼒的Hheuristic启发式的;启发法;启发程序hone怀恋;磨hyperplane超平⾯Linitial最初的implement执⾏intuitive凭直觉获知的incremental增加的intercept截距intuitious直觉instantiation例⼦indicator指⽰物,指⽰器interative重复的,迭代的integral积分identical相等的;完全相同的indicate表⽰,指出invariance不变性,恒定性impose把…强加于intermediate中间的interpretation解释,翻译Jjoint distribution联合概率Llieu替代logarithmic对数的,⽤对数表⽰的latent潜在的Leave-one-out cross validation留⼀法交叉验证Mmagnitude巨⼤mapping绘图,制图;映射matrix矩阵mutual相互的,共同的monotonically单调的minor较⼩的,次要的multinomial多项的multi-class classification⼆分类问题Nnasty讨厌的notation标志,注释naïve朴素的Oobtain得到oscillate摆动optimization problem最优化问题objective function⽬标函数optimal最理想的orthogonal(⽮量,矩阵等)正交的orientation⽅向ordinary普通的occasionally偶然的Ppartial derivative偏导数property性质proportional成⽐例的primal原始的,最初的permit允许pseudocode伪代码permissible可允许的polynomial多项式preliminary预备precision精度perturbation 不安,扰乱poist假定,设想positive semi-definite半正定的parentheses圆括号posterior probability后验概率plementarity补充pictorially图像的parameterize确定…的参数poisson distribution柏松分布pertinent相关的Qquadratic⼆次的quantity量,数量;分量query疑问的Rregularization使系统化;调整reoptimize重新优化restrict限制;限定;约束reminiscent回忆往事的;提醒的;使⼈联想…的(of)remark注意random variable随机变量respect考虑respectively各⾃的;分别的redundant过多的;冗余的Ssusceptible敏感的stochastic可能的;随机的symmetric对称的sophisticated复杂的spurious假的;伪造的subtract减去;减法器simultaneously同时发⽣地;同步地suffice满⾜scarce稀有的,难得的split分解,分离subset⼦集statistic统计量successive iteratious连续的迭代scale标度sort of有⼏分的squares平⽅Ttrajectory轨迹temporarily暂时的terminology专⽤名词tolerance容忍;公差thumb翻阅threshold阈,临界theorem定理tangent正弦Uunit-length vector单位向量Vvalid有效的,正确的variance⽅差variable变量;变元vocabulary词汇valued经估价的;宝贵的Wwrapper包装分类:。
新教材译林版英语选择性必修第三册UNIT 1UNIT 1高频词汇知识点清单Unit 1 Wish you were here高频词汇精讲1. distinct adj.截然不同的,有区别的,不同种类的;清楚的,清晰的;确切的It is also one of the most ethnically diverse nations in the world with distinct cultures.它也是世界上最具种族多样性的国家之一,有着丰富多样的文化。
(教材P2)情景导学The two suggestions are quite distinct from each other.这两个建议截然不同。
用法归纳be distinct from ...与……不同词汇拓展①distinguish v.区分;使出众distinguish A from B/between A and B 把A和B区分开distinguish oneself 享有盛名•At what age are you able to distinguish between right and wrong?你到什么年龄才能明辨是非?②distinguished adj.著名的;卓越的;有尊严的be distinguished for...因……而著名•Although Yuan Longping has passed away, we all know his distinguished achievements in the area of hybrid rice.虽然袁隆平去世了,但我们都知道他在杂交水稻领域的杰出成就。
(写作—人物介绍)③distinctly adv.明显地;清楚地;确切地④distinction n.区别;卓越,优秀;不同凡响•But distinctions that were sharp to us as children become unclear.但是,对小时候的我们来说很明显的区别变得模糊了。
New WordsStrategy 策略Means 方法Diligence 勤勉diligent Prolonged 持续很久的prolong vt延长Nevertheless ad 然而Command 掌握Sustained 持久的sustain 保持Helpful 有帮助的Complain vi 抱怨complant Memorize vt 记住Cram vt 把塞满Bound 一定的Constant 经常发生的Commit vt 把托付给Acquaintance 了解Concentrate v 专心Effective 有效的Route 路线Enlarge v 扩大V ocabulary 词汇Idiomatic 合乎语言习惯的idiom 习语Usage 惯用法Basis 做事的方法Addition 增加Moive 电影Repetition 重复Opportunity 机会Communicate 交流情况Enjoyable 令人愉快的Campus 校园Rehearse vt 排演Rehearsal Partner 同伴Instance 事例Detail 详情Purchase 购买Environment 环境Reliable 可靠的Source 来源Input 输入Rely vi 依赖Handle vt 处理Apart 相隔Assign 分配任务给某人assignment 学生的作业分配的任务Pal 朋友朋友Motivation 动机激起使有动机 激起动机 motivate vt 使有动机Culture 文化文化Summarize vt 总结总结Process 过程过程Accumulation 积累积累积累 accumulate 积累Absorb 吸收吸收Phrases &EapressionsBy no means 决不决不决不At fault 有过错有过错At a time 每次每次每次Be bound to 肯定会肯定会肯定会Commit to memory 记住记住记住Watch out for 密切注意密切注意密切注意Learn of/about 得知得知On a regular/daily/weekly basis 经常每周每天 每周经常 每天In addition to 除外Over and over again 反复反复Seek out 寻找寻找Feel freee 随意随意For instance 例如例如In detail 详细的详细的Act out 将表演出来表演出来Apart from 除外Put into practice 把付诸实践付诸实践。
大学英语四六级考试考前突击必备的1000个核心单词适用于:英语六级、英语四级、英语专四1、prosecute [ 'prosikjuit]vt.对…提起公诉,告发,检举;继续从事w.起诉,【记】词根记忆:pro(向前、大量的)+secut(跟随)+e:大量跟随:该被检举了!向前跟随:继续从事2、transition [traen'zijan] n.过渡,转变区分transit 运输区分transient 转瞬即逝的3、casualty [ 'kaesuslti] n.伤亡人员,受害人【记】来自casual (a办偶然的):偶然遇害了4、permeate [丨psimieit] v.弥漫,遍布,散布;渗入,渗透【记】词根记忆:per(贯穿)+meat+e 空气中贯穿着肉味:弥漫!5、pendulum [ 'pendjubm] n.摆,钟摆;摇摆不定的事态(或局面)【记】词根记忆:pend(悬挂)+ulum —>•摇摆6、refund [ 'rilfAnd] n.退款[rii'fAnd] vt.退还(钱款)【记】词根记忆:re(向后)+fund(资金)—•退回资金—退款7、linger [ Woga] w.(因不愿离开而)留恋徘徊;慢消失【记】联想记忆:歌手(singer)留恋(linger)曾经的舞台【短语】linger on继续存留8、converge [k3niv3:d3] w.(在一点上)会合,互相靠拢;聚集,集中;(思想、观点等)十分相似,相同【记】词根记忆:con(共同)+verge(转)—转到一起—聚集【记】联想记忆:in+cent(分,分币)+ive —用钱(分币)刺激—刺激,鼓励9、associate [saujieit] n•伙伴,同事v.(在思想上)把…联系在一起;使联【短语】associate with结交,交往10、royalty [ 'roblti] n.王族(成员);[fpi ](著作的)版税【记】【记】来自royal(王族的)+ty —王族(成员)区分loyalty 忠诚11、dentist [ 'dentist] n.牙科医生【记】词根记忆:dent(牙齿)+ist(人)—牙科医生12、literary [ 'litarsri] adj.文学(上)的;文人的,书卷气的【记】词根记忆:liter(文字)+ary —文字上的—文学的13、transmit [traenz'mit] vt.传送,传递,传染;播送,发射【记】词根记忆:trans(穿过)+mit(送)—送过去—传送14、mock [mok] v.嘲笑,嘲弄;15、【考】aviation [leivi'eijan] n.航空,航空学;飞机制造业16、【考】incentive [in'sentiv] n.刺激,鼓励17、【考】overwhelm [isuva'welm] vt.(感情上)使受不了;征服,【记】词根记忆:avi(鸟)+ation(表名词)—像鸟一样飞—•区分navigation 导航18、reliable [rHaiabl]邮•.可靠的,可依赖的区分liable 易于……的19、fling [flir)]/2.使投身【记】联想记忆:fl(看作fly,飞)+ing一扔,掷20、anticipate [^n'tisipeit] vr•预料,预期,期望;先于…行动,提前使用【记】词根记忆:anti(先)+cip(拿)+ate(做)—►先拿到—►先于…行动’21、rupture [破裂,断裂;(关系的)决裂,断绝(使)破裂【记】词根记忆:rupt(断)+ure —>断裂,破裂22、gleam [gliim] n.闪光,闪亮;闪现,流露v/.闪烁【考】gleam with流露23、salvation [sael'veijsn] n.拯救,解救;解救措施【记】联想记忆:salv(看作save,拯救)+ation(表动作)•拯救24、pedestrian 行人25、interaction 相互干扰26、【多次考】faculty 才能、能力27、clutch 控制、企图抓住、离合器【短语】clutch at 企图抓住28、mourn 哀悼29、limp 跛行区分limb 腿(climb爬山,爬山要用腿)30、【考】novelty 新奇31、profitable 有利可图的32、jerk 猛拉音似:这客——这客机,猛拉!33、hypothesis 假设34、limb 腿区分climb 爬;limp 跛的用limb去climb,爬下(p)了,说明腿是跛的(limp)35、discipline 纪律、【考】学科36、magistrate 地方法官37、burial 埋葬来自bury38、oxygen 氧气39、【考】simultaneous [~siməlˈteinjəs] 同步的40、attach 使附属41、tolerant 容忍的音似:逃了忍他变形:【考】tolerance 容忍42、squad 小组各个小组(squad)在广场(square)上讨论43、proficiency 熟练、精通pro大量+fic做=大量做——精通44、sentiment 意见一“三”他他就踢门显然有意见45、gear 工具、适应有工具就适应多了46、literacy 识字47、agreeable 令人愉快的48、sticky 黏性的(天气)湿热的49、marsh 沼泽湿地Mars(火星)——火星上都是沼泽50、patron [ˈpeitrən] 资助人patr(父亲)——像父亲一样对待——资助人51、【多次考】reinforce 加强rein缰绳+ force强迫= 为加强训练用缰绳来强迫52、dub 给……起外号;译制影片dub本身为词根意为双;绰号就是第二个名字;译制影片就是影片的第二代53、community 社会54、dazzle 使炫目55、【考】outfit 团体56、heel 脚后跟57、romance 恋爱58、dispatch 公文、急件、派遣dis分开+ patch(片)= 把人群分成几片去派遣急件59、flutter 紧张、飘扬fl(看做fly,飞)+ utter(butter看做蝴蝶butterfly)——放飞蝴蝶,第一次飞很紧张、想飘扬60、rash 轻率的;皮疹61、stitch 一针;缝死地吃——患者都要死在地下了,你还在这吃??快给患者伤口缝一针!62、【考】competitive 竞争力63、corpse 尸体64、fiction 小说65、straightforward 正直的66、spite 恶意spy是间谍间谍普遍怀有恶意67、luxury 奢侈的68、accommodate 容纳提供住宿69、primarily 首要的70、setback 挫折set移动+back后退遇到挫折总是有人想向后移动怯懦不前71、restrain 阻止遏制72、gossip 流言蜚语go进行+s+sip吸吮流言蜚语总是能吸吮去人们的精力73、reward 报答74、【考】maintain 维持75、【考】define 下定义76、harmony 和谐77、【考】arrogant 傲慢的ar加强+rog要求+ant(……的)——一再要求——傲慢的78、paradise 天堂79、flat 平的80、ponder 思索pond 重量通过思考掂掂这件事的重量81、ignite 点燃读音记忆:一搁奶他东西往上一搁他奶奶腿的着火了!------点燃!82、asset 财产区分:assess 给(财产)估价as不断+ set 放置————不断放置财产83、decent体面的84、recipe【`resipi】烹饪区分receipt 【ri`si:t】发票85、premier 首相86、casual 偶然的87、severe 严重的88、propel 推动89、authorize 授权90、lure 诱惑pure纯、lure诱惑————纯属诱惑91、genuine 真正的92、【考】consensus 一致同意93、artery 干线动脉联想art——文艺新干线94、internal 内在的95、splash 飞溅96、spoil 溺爱战利品97、sponge 海绵【短语】sponge off 依赖他人生活海绵就像个死胖子98、hostile 地对的99、chord 心弦和弦100、feeble 虚弱的fee 钱需要花钱看病的101、spontaneous 自发的死胖子忒呐死死胖子都是自发的区分:propaganda 宣传sponge 海绵102、spouse 配偶103、sparkle 闪光spark 星花火闪光104、receipt 【ri`si:t】发票区分recipe【`resipi】烹饪】40、compliment 赞美105、poverty 贫穷106、scandal 丑闻scan 扫描扫描时事,揭露丑闻107、overthrow 推翻108、repression 压抑109、abstract 摘要abs离去tract拉————把大意从文章中拉出来110、ensue 紧着发生ensure确定的事就会接着发生ensue111、【考】cognitive 认知的112、lubricate 使润滑撸不睿——因为有润滑剂的诞生硬撸就不睿智了(男生懂的~)113、notion 概念no不用+ tion读音审——书上的概念不用审114、sprinkle 少量;洒spring 春天春天的阳光洒在身上很舒服115、【考】acute 急性的a+cut切得了急性阑尾炎马上切除手术116、spy间谍【短语】spy on 暗中监视117、rectify 纠正rect直的+ ify(使……)=使他直——纠正118、overhead 在头顶上、管理费用管理费用昂贵,这个烦恼一直在头顶上挥之不去119、destiny 命运、定数120、cellar 地下室cell密室121、intricate [ˈintrikət]错综复杂的人吹可他——人家吹牛你就可着他吹吧,这世界多么错综复杂啊!122、submarine 潜水艇sub(在……下面)+ marine(海洋)——行驶在海洋下,潜水艇123、【考】bureaucracy [bjuəˈrɔkrəsi]官僚主义124、【考】liability 责任lia说谎+ ability能力——谎称自己有能力是不负责任的行为125、escort [ˈeskɔ:t]护送126、subordinate [səˈbɔ:dinəit]下级;处于重要地位【短语】subordinate to 次要的、从属的127、 junk 破烂;丢弃【短语】junk mail 垃圾邮件128、in abundance 充足129、submit to 屈服130、submit oneself to 遵守131、different between 区分;区别132、divert [daiˈvə:t] 使转向133、【考】spiral 螺旋的134、terminate 终止termin(结束)135、render 致使rend(rond,圆的)+ er(ear耳朵)——圆耳朵,致使他被当做怪物136、eagle 鹰137、encourage 鼓励138、【考】conspicuous 显眼cons(引起共同)+ spic(speak,说)——引起共同说:你的过分显眼,引起了大家的讨论!139、prose 散文p(扑)在rose(玫瑰)上,浪漫的感觉更能写出情书般的散文140、pretext 借口、托词per(预先)+ text(写好的文字)——理由竟然是事先就写好的!!!显然他在有预谋的找借口、找托词!142、【考】discrimination 歧视143、【考】rivalry 竞争144、glamour 魅力过来摸——有魅力145、nourish 养育、怀有(希望等)146、stab 刺【短语】stab in the back 背后中伤、背叛147、armor 盔甲arm武器+ or物——保护物148、enlighten 启发en使+ lighten 点亮149、idiom 习语150、progressive 渐进的pro大量的——大量挑衅需要循序渐进151、booth 货摊b+故事——不是故事是真报纸——卖报纸的货摊152、appendix 阑尾、附录ap加强+pend悬挂=悬挂在书后的是附录153、corrode 腐蚀=corrupt154、groove 沟、槽区分groom 准备、新郎想象那个图155、owe 欠156、stability 稳定157、【考】stabilize 使稳定158、flare 照明弹、突然发怒古代用fire照明,如今发展成flare159、stable 马厩160、innumerable 无数的161、justify 证明……是正当的just正义+ ify使162、propaganda 宣传Prop支持者+ agenda看做agenda议程——支持者宣传议程163、foam 泡沫164、compose 组成165、psychiatrist 心理学家166、costume 礼服167、reaction 反应区分react反应、reflect反映、reflex反射168、presume 假定【短语】presume on 不正当利用169、entail 使承担en给+ tail尾巴——给一个尾巴用以承担170、referee 裁判员171、dodge 托词、回避总是找托词回避,像狗一样172、stack 堆库存(stock)一堆(stack)商品,堆到一起对大家不利:【考】对……不利【短语】stack against 对……不利173、circus 马戏团174、disguise 伪装迪斯尼用东西盖着肯定要伪装什么175、stagger 蹒跚Stag看做stay停——停不稳——蹒跚176、recognize 认出(o后面易多写n,要注意)177、staircase 楼梯178、embody 使具体化179、artillery 大炮180、【考】dwell 居住d(do,做)+ well(好)——做得好就要留下来居住【考】【短语】dwell on/upon 老是想着,详述181、parameter 界限、参数para(辅助)+ meter(测量)——辅助测量的,是参数,参数一般都是有界限的区分paralyze 使瘫痪182、scramble 爬行抢夺183、imaginary 想象的184、relief 宽慰区分belief 相信185、allocate 分派al(加强)+ loc(地方)+ ate(做)——分派到各地去做186、compulsory 强制性的区分capsule 胶囊187、flask 长颈瓶188、【考】expertise 专长189、terrace 地棚terr(地)+ race(比赛)——比赛的地方:地棚区分terrain 地形terr(地)+ rain 下雨——下雨与地形有关190、loyalty 忠诚191、sieve 筛选、滤192、recur 再发生re(再)+ cur(发生)——再发生区分incur 招致in(进来)+ cur(发生)——进来发生:招致193、【考】capable 有能力的区分【考】capability 能力;功能区分【考】capacity 能力;容量区分【考】facility 设备194、【考】mediate 调节195、ammunition 弹药;军火am(加强)+ mum(公共的)——军火是保卫公共安全的基础196、critical 批评的197、【考】massive 大量的198、decree 判决和degree(度,刑度)一起记忆199、cucumber 黄瓜200、terrain 地形terr(地7、)+ rain 下雨——下雨与地形有关201、robust 强壮的像机器人一样强壮202、saturate 使饱和203、【考】scatter 驱散音似:死该他!——死是他活该的!这种人你还不驱散他?区分【考】scanty 缺乏s(死)+ can(能)+ 踢——死了才能会踢——言外之意一辈子也不会踢球,这真缺乏信心啊!204、defect 缺点de(变坏)+ fect(做)——往坏了做:缺点!205、convict 囚犯、宣判有罪206、ingredient 成分207、retort 反驳re(往回)+ tort(教)——往回教:学生教老师——学生反驳了老师的学术观点208、brief 摘要区分relief 快慰;belief 信任209、pasture 牧场210、【考】terrific 极妙的211、purchase 买212、mobilize 动员213、assassination 暗杀as(加强)+ sas(看做SAS 非典)+ sin(son儿子)——非典暗杀了很多孩子214、detect 察觉215、【考】mutual 相互的216、preach 宣讲、竭力鼓吹p(屁)+ reach(到达)——屁老师马乾凯到了——又要开始宣讲、吹牛了217、territory 领土terr(地)+ iory(地点)——领土区分terrain 地形区分terrace 地棚218、insight 洞察力219、desirable 值得向往的220、instrumental 起作用的、用乐器演奏的来自instrument 手段、器具221、testify 证明222、【考】gamble 赌博222、strand 使搁浅死船的——死在船上的——搁浅了223、concise 简明的con(共同的)+ cise(切掉)= 把相同的套话都切掉,留下主干部分——简明的224、strategy 战略225、concept 概念226、forge 仿制发音记忆:仿制227、testimony 见证228、certainty 必然的事229、detain 拘留230、fantastic 极好的231、texture 质地232、acquaintance 熟人233、migrant 移民者234、mingle 使混合各种single(单一的)在一起就mingle(混合)了【短语】mingle with 与……相融合235、theft 偷窃236、mutter 轻声耳语mutter说butter 轻声要黄油!(半夜饿了,不好意思大声说)237、dissipate 使消散dis(不)+ sip(小口喝)+ ate(吃)——不小口吃喝——大吃大喝——浪费、挥霍、使消散!238、profession 职业239、financial 财政的240、denounce 谴责de(变换)+ nounce(报告)——弄坏报告、要受谴责!241、catholic 天主教的cat(猫,教堂里把猫看得神圣)+ holi(看做holy,神圣的)= 天主教242、streak 疾驰、条纹死醉渴——都醉死了、拿奶奶的带条纹的手巾擦嘴,奶奶见状疾驰过来告诉我不许动脏她的毛巾243、streamline 使成流线型244、maneuver 调度723动车事故的时候,调度员还在那摸牛玩呢,结果信号给错了!245、alleviate 缓解阿李伟他——李伟是数分老师,他太胖了,走路需要缓解246、physiological [.fiziəˈlɔdʒikəl]生理的区分physician 医生247、facet 一个方面248、stride 大步大步走和骑车(ride)一样快区分ride 骑车249、【考】apparatus 仪器ap(加强)+ para(辅助)+ tus——起辅助作用——仪器250、rival 竞争对手251、deduce 推断弟丢死——弟弟丢了死了——推断出来的252、levy 税款来喂——交税给国家,国家干部等着老百姓来喂253、disposition 性格dis(分开)+ position(位置)——因为性格不合所以分开位置254、scheme 计划、方案255、periodical 时期的、期刊256、exhaust 疲倦的、排气装置疲倦需要排气257、stereotype 定式258、stern 严厉的死的按除非死了否则谁也不许在训练者按动暂停按钮——太严厉了!259、census 统计、人口普查260、repay 归还261、independent 独立262、giggle【'gigl】咯咯笑263、glitter 闪光g + litter(看做little,小)——一闪一闪小星星264、lieutenant [lefˈtenənt;lu:ˈtenənt] 陆军中尉海军上尉265、evil 祸害266、prospective 预期的Pro向前+ spect看——向前看的——预期的267、dismiss 解雇268、endurance 忍受力269、conserve 保存270、expense 花费【短语】at the expense of 以……为代价271、preclude 排除Pre提前+ clude推断——提前推断出故障、排除险情区分conclude 推断272、compete 比赛273、bargain 便宜货274、【考】external 外部的275、aspiration 志向276、drawback 缺点277、cute 精明的278、toll 过路费;死亡人数区分tool交通工具交通工具肯定要交过路费交通工具出事了要有死亡人数279、correlate [`krəlet] 关联【短语】correlate to/with 关联280、sip 小口喝281、conservation 保存282、clasp 扣子;抱紧283、current 电流、潮流区分currency 货币284、invert 使倒转285、segment 部分286、religious 宗教的287、extraordinary 非同寻常的extra额外的+ ordinary普通的——非同寻常的288、drain 排水沟;慢慢减少d(堵)+rain(雨)——堵雨的——排水沟、雨水进排水沟会慢慢减少289、confidence 信任290、【考】strain 拉紧291、entrepreneur [.ɔntrəprəˈnə:]企业家来自enterprise 事业292、assault 攻击as(加强)+ sault(看做salt盐)——往伤口上撒盐——攻击293、【考】encounter 邂逅遇到294、【考】elementary 基本的、刚具备要求的295、【考】fertilizer 肥料词源:【考】fertile 富有的296、aggregate 总数【短语】in the aggregate 作为总体,总共297、eliminate 消灭298、confident 自信300、rank 军衔、【考】排名301、hazard 危险302、band 乐队303、coarse 粗糙的音似:靠死——靠!粗糙死了!304、justification 正当的理由305、command 命令区分demand 要求306、thrill 兴奋307、endeavor 努力烟袋——努力戒烟308、【考】thrive 兴旺309、shield 防护物310、invisible 看不见的311、interact 相互作用312、marshal 最高指挥官;【考】安排马首——最高指挥官:负责安排!313、throne [θrəun]王权thr(看做the)+ one——为我独尊:王权!区分thorn [θɔ:n] 荆棘314、guardian 监护人315、meadow 草地mea(看做meal餐)+ dow(音似:斗,斗牛)——斗牛的时候给牛吃的早餐:草地区分:pasture 牧场316、provoke 挑衅317、tick 打”√”318、tickle 逗乐,发痒买张ticket(电影票)让她tickle(逗乐)一下区分tackle 与……交涉319、indefinite 无限期的变形:【考】indefinitely 无限地320、restore 归还321、predominant322、obscene 淫秽的ob(反)+ scene(景色)——反能量的景色:淫秽图像323、criterion 准则324、queer 奇怪的、不舒服的排队(queue)这么久竟然买不到票,真奇怪(queer)、心里不舒服325、liberal 自由的326、theme 主题、题目327、therefore 之后328、mistress 主妇329、promising 有希望的330、therefore 因此331、parasite 寄生虫para(辅助)+ sit(坐在、居住)——辅助住在其他动物里面的——寄生虫区分paralyze 使瘫痪区分parameter 参数、界限332、famine 饥荒、严重的缺乏音似:发米呢!饥荒了得发米!333、militant 激进的音似:密里探他:秘密探访他,防止有激进的行为造成伤害334、dilute 冲淡、稀释di(die死)+ lute(溜他)——人死了却溜了,肯定是杀人者逃逸了,把血液稀释冲淡,防止被抓!335、slaughter 屠杀336、arc 弧337、conference 会议338、consequently 因此区分consequence 后果339、graduate 毕业生340、solo 独奏341、thermal 热的、保暖的音似:傻帽!不知道保暖,你傻帽啊!难道是热的?342、inaugurate 为……举办开幕式音似:一闹就睿他——开幕式嘛,闹闹哄哄的办才睿智呢!太死板就没气氛了!343、【考】diverse 相异的di(不同、区分)+ verse(转)——往不同方向转——相异的!区分inverse、converse 转化344、drought 干旱dr(dry,干的)+ ought(应该)——因为干旱,所以一切都很干这是应该的!345、thesis [ˈθi:sis] 论文346、optimism 乐观v、【考】optimize 优化347、flaw 缺点、瑕疵f(飞离)+ law(法律)——背离的法律:太严重的缺点、瑕疵了!348、thigh [θai]大腿音似:晒大腿怕晒!349、smash 击碎350、thorn 荆棘t+ horn(角)——荆棘351、brood 一家的孩子、一窝、沉思一家的孩子(brood)流着相同的血(blood),家人一起陷入沉思352、outbreak 突然发生353、compensate 弥补354、excursion 远足ex(出去)+cursion(磕神:出去给神磕头去——远足:去歪脖老母那儿,哈哈)355、dedicate 把(一生)等献给音似:dedi(爹地)+ ate(做)——爸爸还有做的事情——一生献给的事业356、【考】confront 遭遇357、terrify 使害怕terr(地)+ 飞——从地面飞起来——害怕!区分terrific 极好的(fic是做的意思,做的好!——极好的)区分terrace 地棚区分terrain 地形区分territory 领土358、chunk 相当大的部分(或数量)359、assistance 援助v。
EnhancedTreeClusteringwithSinglePronunciationDictionaryforConversationalSpeechRecognition
HuaYuandTanjaSchultzInteractiveSystemsLab,CarnegieMellonUniversity,Pittsburgh,PA15213hyu@cs.cmu.edu
AbstractModelingpronunciationvariationiskeyforrecogniz-ingconversationalspeech.Ratherthanbeinglimitedtodictionarymodeling,wearguethattriphoneclusteringisanintegralpartofpronunciationmodeling.Weproposeanewapproachcalledenhancedtreeclustering.Thisapproach,incontrasttotraditionaldecisiontreebasedstatetying,allowsparametersharingacrossphonemes.Weshowthataccuratepronunciationmodelingcanbeachievedthroughefficientparametersharingintheacous-ticmodel.Combinedwithasinglepronunciationdic-tionary,a1.8%absoluteworderrorrateimprovementisachievedonSwitchboard,alargevocabularyconversa-tionalspeechrecognitiontask.
1.IntroductionModelingconversationalspeechisamajorchallengeforcurrentspeechrecognitionresearch.Conversationalspeechischaracterizedbyrampantpronunciationvari-ations,whereaccuratepronunciationmodelingcanleadtohighrecognitionperformance.Traditionally,peoplehavetriedtoaddalternativepronunciationstotherecog-nitiondictionary.Despiteextensiveinvestigation,thishasyieldedonlymarginalimprovement.Whatispronunciationmodeling?Ispronunciationmodelingasynonymfordictionarymodeling?AsshowninFigure1,theboundarybetweenpronunciationmod-elingandacousticmodelingisnotclear.Foragivenword,thelexiconisfirstlookeduptoconvertthewordintoaphonemesequence,whichissubsequentlytrans-latedintoastatesequenceusingaphoneticdecisiontree.Thestatesequenceisultimatelyusedtoalignwiththeacousticobservation.Hencethephoneticdecisiontree,atraditionallyacousticmodelingconcept,alsoplaysakeyroleinthemappingfromsymbolic(phoneme)leveltomodel(state)level.Subtlepronunciationvariationsmayactuallybebet-termodeledimplicitlyattheacousticmodellevel,ratherthanmodeledexplicitlyatthelexicallevel.Forexample,phonemeAXcanbealternatelyrealizedasphonemeIXincertainwords:
AFFECTIONATEAXFEHKSHAXNAXTAFFECTIONATE(2)AXFEHKSHAXNIXT
PronunciationModel
wordsequence
phonesequencestate/modelsequence
pronunciationlexicon
phoneticdecisiontree
Figure1:Pronunciationmodelasamappingfromsym-bolicleveltomodellevel
Insteadofaddingavariantinthedictionary,wecankeepthedictionaryunchanged,andeitheraugmentthemixturemodelofAXwithGaussiansfromthemixturemodelofIX(asin[1]),orsimplytiethesetwomodelstogether.Thispaperfocusesonimplicitpronunciationmod-elingusingdecisiontreebasedtying.InSection2,wewillfirstexaminetherelativemeritsofdifferentpronun-ciationmodelingmethods,namely,explicitlyaddingdic-tionaryvariantsversusimplicitmodelingusingdecisiontreebasedstatetying.Section3introducesenhancedtreeclusteringthatallowsparametersharing/tyingacrossdifferentphonemes.ThenewapproachisevaluatedonSwitchboard,alargevocabularyspeechrecognition(LVCSR)task.Section4presentsexperimentsanddis-cussions.RelatedresearchworkisreviewedinSection5.
2.Explicitvs.ImplicitPronunciationModeling
Earlyattemptsatpronunciationmodelingtaketheformofmanuallyeditingalexicon.Forexample,tomodeltheflappingofTinBETTER,analternativelexicalentryBETTER(2)isintroduced:
BETTERBEHTAXRBETTER(2)BEHDXAXR
Interestingly,forthiskindofpronunciationvariation,triphonemodelingturnsouttobebetterthandictionaryediting.Triphonewasoriginallyintroducedtomodelcontextdependency.Asthereisalargenumberoftri-phonesinanLVCSRsystem,decisiontreebasedstatetyingiswidelyusedtoclustertriphones[2].Thisensuressufficienttrainingdataforeachmodelaswellasbetter
EUROSPEECH 2003 - GENEVA1869generalizationtounseencontexts.Itturnsoutthatmanycoarticulationrules,suchastheflappingofT,canbewellcapturedbytriphonemodelsinapurelydata-drivenfash-ion.Thisgivesusanothersolution:leavethedictionaryunder-specified,anduseautomatictriphoneclusteringforpronunciationmodeling1.Theuseofmixturemodelastheunderlyingdistributionisalsoimportant,sinceweareusingthesametriphonemodelforboththeflappedandunflappedversionofT.Comparedtotriphoneclustering,manuallyeditingalexiconisbothlaborintensiveanderrorprone.IntheexampleoftheflappingofT,oneneedstobeextremelycarefultomakesurethatallrelevantdictionaryentriesaremodified,whilenothingelseiserroneouslychanged.Thisturnsouttobequitedifficultinreality.Whereasintheautomaticsolution,itisaloteasiertokeepthedictionarysimpleandconsistent.Onecouldalsouseautomaticprocedurestogener-atepronunciationvariants,inordertoavoidthepitfallofmanualediting.Therehasbeenalotofresearchinthisarea.Sofar,theimprovementhasbeenmarginal.Thiscouldbeattributedtoseveralundesirablesideeffectsofpronunciationvariants:First,addingvariantsincreaseslexicalconfusabil-ityinarecognitiondictionary;Second,ifnotdoneproperly,addingvariantsin-creasesmodelconfusabilityduringtraining.Asil-lustratedinFigure2,whenavariantreplacesphoneAbyphoneB,wearedistributingtomodelBthedatathatwasoriginallyusedtotrainmodelA.Incaseswherethevariantisspurious,modelBwillbecontaminatedwithdatabelongingtoA,makingAandBmoreconfusable.variantsB’strainingdataA’strainingdatamodelAmodelBFigure2:ModelContaminationInsummary,duetothecomplexinteractionbetweenlexiconandacousticmodeling,addingpronunciationvariantsshouldbeexercisedwithgreatcare.Inthenextsection,weintroduceenhancedtreeclusteringforim-plicitpronunciationmodeling.3.EnhancedTreeClusteringDecisiontreebasedstatetyingallowsparametershar-ingatleafnodesofatree.Typically,onedecisiontree1Thisactuallyresonateswithsinglepronunciationdictionary[3]isgrownforeachsub-state(begin/middle/end)ofeachphone.With50phonemesinthephoneset,150sepa-ratetreesarebuilt(Figure3(a)).Parametersharingisnotallowedacrossdifferentphonesorsub-states.Withen-hancedtreeclustering,asingledecisiontreeisgrownforallsub-statesofallthephones(Figure3(b)).Theclus-teringprocedurestartswithallpolyphonesattheroot.Questionsareaskedregardingtheidentityofthecen-terphoneanditsneighboringphones,plusthesub-stateidentity(begin/middle/end).Ateachnode,thequestionthatyieldsthehighestinformationgainischosenandthetreeissplit.Thisprocessisrepeateduntileitherthetreereachesacertainsizeoraminimumcountthresholdiscrossed.Comparedtothetraditionalmultiple-treeap-proach,asingletreeallowsmoreflexiblesharingofpa-rameters.Anynodescanpotentiallybesharedbymulti-plephones,aswellassub-states.